1. Nathan Brunelle Department of Computer Science University of Virginia www.cs.virginia.edu/~njb2b/theory Theory of Computation CS3102 – Spring 2014 A tale of computers, math, problem solving, life, love and tragic death

2. Decidability Def: A language is Turing-decidable iff it is exactly the set of strings accepted by some always-halting TM. w→ Input √ Accept & halt  Reject & halt Never runs forever Theorem: The regular languages are decidable. Theorem: The context-free languages are decidable. Theorem: The finite languages are decidable.

3. A “Simple” Problem Let S = {x 3 + y 3 + z 3 | x, y, z  ℤ } Q: Is S infinite? A: Yes, since S contains all cubes. Q: Is S Turing-recognizable? A: Yes, since dovetailing TM can enumerate S. Q: Is S Turing-decidable? A: Unknown! Q: Is 29  S? A: Yes, since 3 3 +1 3 +1 3 =29 Q: Is 30  S? A: Yes, since (2220422932) 3 +(-2218888517) 3 +(-283059965) 3 =30 Q: Is 33  S? A: Unknown! Theorem [Matiyasevich, 1970]: Hilbert’s 10 th problem (1900), namely of determining whether a given Diophantine (i.e., multi-variable polynomial) equation has any integer solutions, is not decidable.

4. Hilbert’s Problems International Congress of Mathematics, Paris, 1900 David Hilbert proposed 23 open problems Most successful open problems compilation ever Set the direction for 20 th century mathematics Hilbert’s problems received much attention to date Several have been resolved (e.g., Continuum hypothesis) Others still open (e.g., Riemann hypothesis) Catalyzed other open problems lists: –Clay Institute’s Millennium Prize problems –DARPA Mathematical Challenges, 2009

5. Introduction from Hilbert’s Lecture “Who of us would not be glad to lift the veil behind which the future lies hidden; to cast a glance at the next advances of our science and at the secrets of its development during future centuries? What particular goals will there be toward which the leading mathematical spirits of coming generations will strive? What new methods and new facts in the wide and rich field of mathematical thought will the new centuries disclose? History teaches the continuity of the development of science. We know that every age has its own problems, which the following age either solves or casts aside as profitless and replaces by new ones. If we would obtain an idea of the probable development of mathematical knowledge in the immediate future, we must let the unsettled questions pass before our minds and look over the problems which the science of today sets and whose solution we expect from the future. To such a review of problems the present day, lying at the meeting of the centuries, seems to me well adapted. For the close of a great epoch not only invites us to look back into the past but also directs our thoughts to the unknown future. The deep significance of certain problems for the advance of mathematical science in general and the important role which they play in the work of the individual investigator are not to be denied. As long as a branch of science offers an abundance of problems, so long is it alive; a lack of problems foreshadows extinction or the cessation of independent development. Just as every human undertaking pursues certain objects, so also mathematical research requires its problems. It is by the solution of problems that the investigator tests the temper of his steel; he finds new methods and new outlooks, and gains a wider and freer horizon. It is difficult and often impossible to judge the value of a problem correctly in advance; for the final award depends upon the gain which science obtains from the problem. Nevertheless we can ask whether there are general criteria which mark a good mathematical problem. An old French mathematician said: "A mathematical theory is not to be considered complete until you have made it so clear that you can explain it to the first man whom you meet on the street." This clearness and ease of comprehension, here insisted on for a mathematical theory, I should still more demand for a mathematical problem if it is to be perfect; for what is clear and easily comprehended attracts, the complicated repels us. Moreover a mathematical problem should be difficult in order to entice us, yet not completely inaccessible, lest it mock at our efforts. It should be to us a guide post on the mazy paths to hidden truths, and ultimately a reminder of our pleasure in the successful solution.”

6. Hilbert’s Problems Problem 1: The continuum hypothesis (conjectured by Georg Cantor: there is no set whose cardinality is strictly between those of the integers and the reals) Status: The continuum hypothesis was proven by Gödel (1939) and Cohen (1963) to be independent of (i.e., impossible to prove or disprove) Zermelo–Frankel set theory. Related open questions remain (e.g., regarding the generalized continuum hypothesis), and there is still much active research in this area. Problem 2: Prove the axioms of arithmetic are consistent. Status: Gödel (1931) proved that the consistency of Peano arithmetic can not be proven within Peano arithmetic itself. Gödel also proved that every consistent formal axiomatic system is incomplete. Gentzen (1936) proved the consistency Peano arithmetic within a different system (that is weaker than set theory).

7. Hilbert’s Problems Problem 6: Axiomatize all of physics. Status: Since Hilbert stated this problem in 1900, relativity theory was developed by Einstein (1905 and 1915), as was quantum mechanics by Dirac (1920’s), followed by other more modern approaches, e.g. quantum field theory, the standard model, quantum gravity, and string theory. Hilbert himself made significant contributions to relativity and physics, but his original problem/goal of axiomatizing all of physics remains mostly open. Problem 7: Is a b transcendental, for algebraic a ≠ 0,1 and irrational algebraic b ? Status: Shown to be true by Gelfond and Schneider (1934), even for complex a and b. This proves that, e.g., e  i i are all transcendental. But many similar problems remain open, such as the trancendance (or even the irrationality) of  e, 2 e, or even  e and  e.

8. Hilbert’s Problems Problem 8: The Riemann hypothesis (the real part of any non-trivial zero of the Riemann zeta function is ½) and Goldbach's conjecture (every even number > 2 can be written as the sum of two primes). Status: Both the Reimann hypothesis (1859) and Goldbach’s conjecture (1742) remain open to this day. The Reimann hypothesis has many far-reaching implications in mathematics, logic, and computer science. It was numerically verified for the first ten trillion zeroes, and appears on the Millennium Prize list ($1M bounty) as well as the ARPA Mathematical Challenges List. The Goldbach conjecture was verified for the first 10 18 values. Problem 9: Find most general law of the reciprocity theorem in any algebraic number field. Status: Partially solved by Artin (1924), Takagi & Hasse, and Shafarevich (1948); still some open issues.

9. Problem 10: Find an algorithm that determines whether a given Diophantine (i.e., multi-variable polynomial) equation has any integer solutions. Ex: x 2 +y 2 =z 2 has many integer solutions (Pythagorean theorem, e.g., x=3, y=4, z=5) x 9 +y 9 =z 9 has no integer solutions (corollary of Fermat’s Last Theorem, conjectured in 1637, proved in 1995 by Andrew Wiles) Many attempts at solution & partial results: Emil Post (1944), Martin Davis (1949), Julia Robinson (1950), Hilary Putnam (1959) Hilbert’s Problems

10. Hilbert’s Tenth Problem Theorem [ Matiyasevich, 1970]: Every Turing-enumerable (aka Recognizable) set is Diophantine (i.e., the solutions of some polynomial) Ex: the set of primes coincides exactly with the positive values of this 26-variable polynomial: (k + 2)(1 − [wz + h + j − q] 2 − [(gk + 2g + k + 1)(h + j) + h − z] 2 − [16(k + 1) 3 (k + 2)(n + 1) 2 + 1 − f 2 ] 2 − [2n + p + q + z − e] 2 − [e 3 (e + 2)(a + 1) 2 + 1 − o 2 ] 2 − [(a 2 − 1)y 2 + 1 − x 2 ] 2 − [16r 2 y 4 (a 2 − 1) + 1 − u 2 ] 2 − [n + l + v − y] 2 − [(a 2 − 1)l 2 + 1 − m 2 ] 2 − [ai + k + 1 − l − i] 2 − [((a + u 2 (u 2 − a)) 2 − 1)(n + 4dy) 2 + 1 − (x + cu) 2 ] 2 − [p + l(a − n − 1) + b(2an + 2a − n 2 − 2n − 2) − m] 2 − [q + y(a − p − 1) + s(2ap + 2a − p 2 − 2p − 2) − x] 2 − [z + pl(a − p) + t(2ap − p 2 − 1) − pm] 2 ) as a, b, c, …, z range over the nonnegative integers!

11. Corollary [ Matiyasevich, 1970]: There is a fixed “universal” polynomial P such that for any Turing-enumerable set S there exists an integer n 0 such that S = {w |  x 1, x 2, …, x k  P(n 0,w,x 1, x 2, …, x k )=0 } i.e., there is a fixed polynomial that can “output” any computable set, depending on one parameter. This is an analogue of a universal Turing machine! Hilbert’s Tenth Problem

12. Recognition and Enumeration Def: An “enumerator” Turing machine for a language L prints out precisely all strings of L on its output tape. Note: The order of enumeration may be arbitrary. b$bba$$aa

13. Decidable = Lexicographically Enumerable Theorem: If a language is decidable, it can be enumerated in lexicographic order by some Turing machine. Proof: Start with a Turing machine E that enumerates Σ* in lexicographic order. Let M decide some language L. Every time E prints a string to its tape M will check if that string is in L. If so it prints the string to a second tape. Def: An “enumerator” Turing machine for a language L prints out precisely all strings of L on its output tape.

14. Decidable = Lexicographically Enumerable Theorem: If a language can be enumerated in lexicographic order by some TM, it is decidable. Proof: Start with a Turing machine E that enumerates L in lexicographic order. We construct M which decides L as follows. Let w be the input to M. M will run E as a subroutine. Every time E prints a string s, M will first check if s=w. If so, the M accepts. Otherwise M will check if s>w, if so then M rejects. Def: An “enumerator” Turing machine for a language L prints out precisely all strings of L on its output tape.

15. Decidable = Lexicographically Enumerable Theorem: If a language is recognizable, then it can be enumerated by some Turing machine. Proof: Start with a Turing machine E that enumerates Σ* in lexicographic order. Let M recognize some language L. Every time E prints a string w to its tape M will check if that string is in L “in parallel”. If M accepts w then print w to a second tape. Def: An “enumerator” Turing machine for a language L prints out precisely all strings of L on its output tape.

16. Decidable = Lexicographically Enumerable Theorem: If a language can be enumerated by some TM, then it is recognizable. Proof: Start with a Turing machine E that enumerates L. We construct M which recognizes L as follows. Let w be the input to M. M will run E as a subroutine. Every time E prints a string s, M will first check if s=w. If so, the M accepts. Def: An “enumerator” Turing machine for a language L prints out precisely all strings of L on its output tape.

17. Theorem: The decidable langs are closed under complement. Modify the machine to exchange the accept state for reject state. Closure Properties of Decidable Languages A string w is in the language of Turing Machine T if δ*(ε, q 0, w) = (u,, v) A string w is not in the language of Turing Machine T if δ*(ε, q 0, w) = (u,, v) q accept q reject

18. Theorem: The decidable languages are closed under union. 1.Copy input 2.Run machine 1 on copy of input (never cross $) 1.If accept then move into accept state 3.Erase everything to right of $ 4.Run machine 2 Closure Properties of Decidable Languages abba$abba input copy M1 M2

19. Theorem: The decidable languages are closed under . 1.Copy input 2.Run machine 1 on copy of input (never cross $) 1.If reject then move into reject state 3.Erase everything to right of $ 4.Run machine 2 Closure Properties of Decidable Languages abba$abba input copy M1 M2

20. Theorem: The decidable langs are closed under concatenation. Use a 3-tape machine: Tape 1: save input remember factor attempts Tape 2: tape for Machine 1 Tape 3: tape for Machine 2 Closure Properties of Decidable Languages abba$0 indexinput a abba 1 bba

21. Closure Properties of Decidable Languages Theorem: The decidable langs are closed under Kleene star. Similar to concatenation, but now we also try all possible numbers of concatenations. Tape 1: abba$0$ index input Number of concatenations

22. Here Be Hilbert’s Problems

23. Hilbert’s Problems Problem 1: The continuum hypothesis (conjectured by Georg Cantor: there is no set whose cardinality is strictly between those of the integers and the reals) Status: The continuum hypothesis was proven by Gödel (1939) and Cohen (1963) to be independent of (i.e., impossible to prove or disprove) Zermelo–Frankel set theory. Related open questions remain (e.g., regarding the generalized continuum hypothesis), and there is still much active research in this area. Problem 2: Prove the axioms of arithmetic are consistent. Status: Gödel (1931) proved that the consistency of Peano arithmetic can not be proven within Peano arithmetic itself. Gödel also proved that every consistent formal axiomatic system is incomplete. Gentzen (1936) proved the consistency Peano arithmetic within a different system (that is weaker than set theory).

24. Hilbert’s Problems Problem 3: Given any two polyhedra of equal volume, is it always possible to cut the first into finitely many polyhedral pieces which can be reassembled to yield the second? Status: Proved via counter-example to be impossible by Hilbert’s student Dehn (1901). The Wallace-Bolyai–Gerwien theorem (1807): in 2D this is always possible for polygons of equal areas. Problem 4: Construct all metrics where lines are geodesics. Status: Too vague for a definite answer. Problem 5: Are continuous groups automatically differential groups? Status: Resolved in the negative by von Neumann (1929), Pontryagin (1934), Gleason-Montgomery-Zippin (1950’s), and Yamabe (1953).

25. Hilbert’s Problems Problem 6: Axiomatize all of physics. Status: Since Hilbert stated this problem in 1900, relativity theory was developed by Einstein (1905 and 1915), as was quantum mechanics by Dirac (1920’s), followed by other more modern approaches, e.g. quantum field theory, the standard model, quantum gravity, and string theory. Hilbert himself made significant contributions to relativity and physics, but his original problem/goal of axiomatizing all of physics remains mostly open. Problem 7: Is a b transcendental, for algebraic a ≠ 0,1 and irrational algebraic b ? Status: Shown to be true by Gelfond and Schneider (1934), even for complex a and b. This proves that, e.g., e  i i are all transcendental. But many similar problems remain open, such as the trancendance (or even the irrationality) of  e, 2 e, or even  e and  e.

26. Hilbert’s Problems Problem 8: The Riemann hypothesis (the real part of any non-trivial zero of the Riemann zeta function is ½) and Goldbach's conjecture (every even number > 2 can be written as the sum of two primes). Status: Both the Reimann hypothesis (1859) and Goldbach’s conjecture (1742) remain open to this day. The Reimann hypothesis has many far-reaching implications in mathematics, logic, and computer science. It was numerically verified for the first ten trillion zeroes, and appears on the Millennium Prize list ($1M bounty) as well as the ARPA Mathematical Challenges List. The Goldbach conjecture was verified for the first 10 18 values. Problem 9: Find most general law of the reciprocity theorem in any algebraic number field. Status: Partially solved by Artin (1924), Takagi & Hasse, and Shafarevich (1948); still some open issues.

27. Problem 10: Find an algorithm that determines whether a given Diophantine (i.e., multi-variable polynomial) equation has any integer solutions. Ex: x 2 +y 2 =z 2 has many integer solutions (Pythagorean theorem, e.g., x=3, y=4, z=5) x 9 +y 9 =z 9 has no integer solutions (corollary of Fermat’s Last Theorem, conjectured in 1637, proved in 1995 by Andrew Wiles) Many attempts at solution & partial results: Emil Post (1944), Martin Davis (1949), Julia Robinson (1950), Hilary Putnam (1959) Hilbert’s Problems

28. Hilbert’s Tenth Problem Theorem [ Matiyasevich, 1970]: Every Turing-enumerable set is Diophantine (i.e., the solutions of some polynomial) Ex: the set of primes coincides exactly with the positive values of this 26-variable polynomial: (k + 2)(1 − [wz + h + j − q] 2 − [(gk + 2g + k + 1)(h + j) + h − z] 2 − [16(k + 1) 3 (k + 2)(n + 1) 2 + 1 − f 2 ] 2 − [2n + p + q + z − e] 2 − [e 3 (e + 2)(a + 1) 2 + 1 − o 2 ] 2 − [(a 2 − 1)y 2 + 1 − x 2 ] 2 − [16r 2 y 4 (a 2 − 1) + 1 − u 2 ] 2 − [n + l + v − y] 2 − [(a 2 − 1)l 2 + 1 − m 2 ] 2 − [ai + k + 1 − l − i] 2 − [((a + u 2 (u 2 − a)) 2 − 1)(n + 4dy) 2 + 1 − (x + cu) 2 ] 2 − [p + l(a − n − 1) + b(2an + 2a − n 2 − 2n − 2) − m] 2 − [q + y(a − p − 1) + s(2ap + 2a − p 2 − 2p − 2) − x] 2 − [z + pl(a − p) + t(2ap − p 2 − 1) − pm] 2 ) as a, b, c, …, z range over the nonnegative integers!

29. Corollary [ Matiyasevich, 1970]: There is a fixed “universal” polynomial P such that for any Turing-enumerable set S there exists an integer n 0 such that S = {w |  x 1, x 2, …, x k  P(n 0,w,x 1, x 2, …, x k )=0 } i.e., there is a fixed polynomial that can “output” any computable set, depending on one parameter. This is an analogue of a universal Turing machine! Hilbert’s Tenth Problem

30. Hilbert’s Problems Problem 11: Solving quadratic forms with algebraic numerical coefficients. Status: Partially solved by Hasse (1923). Problem 12: Extend the Kronecker–Weber theorem on abelian extensions of the rational numbers to any base number field. Status: Still unsolved. Problem 13: Solve all 7-th degree equations using functions of two parameters. Status: Partially solved by Kolmogorov (1956), Arnold (1957), and Shimura (1976). Problem 14: Proof of the finiteness of certain complete systems of functions. Status: Counter-examples found by Nagata (1959).

31. Hilbert’s Problems Problem 15: Find a rigorous foundation for Schubert's enumerative calculus. Status: Partially resolved. Problem 16: Topology of algebraic curves and surfaces. Status: Open-ended: some results, but unresolved. Problem 17: Expression of definite rational function as quotient of sums of squares Status: Resolved in the affirmative by Artin (1927) and Delzel (1984).

32. Hilbert’s Problems Problem 18: Is there a non-regular, space-filling polyhedron? What is the densest sphere packing? Status: Anisohedral tilings were found in 3D by Reinhardt (1928), and for 2D by Heesch (1935). Sphere packing in 3D (Kepler’s problem, 1611) was solved by Toth (1953) and Hale (1998). Regular sphere packing in 24 dimensions was solved by Cohn and Kumar (2004), where the “kissing number” is 196,560. Many related open problems remain, including non-regular, non-uniform, and ellipsoid packings.

33. Hilbert’s Problems Problem 19: Are solutions of Lagrangians always analytic? Status: Resolved in the affirmative by Bernstein (1904). Problem 20: Do all variational problems with certain boundary conditions have solutions? Status: Resolved in the affirmative. Problem 21: Proof of the existence of linear differential equations having a prescribed monodromic group Status: Resolved by Plemelj (1908), Schlesinger (1964), Dekkers (1978), and Bolibrukh (1989). Problem 22: Uniformization of analytic relations by means of automorphic functions Status: Resolved. Problem 23: Further development in calculus of variations Status: Unresolved.